SYSTEM AND METHOD FOR PROVIDING IMPROVED CLAIMANT AUTHENTICATION

摘要

A system and method are provided for improving the accuracy of speakerauthentication by combining the results of multiple verification sources usingstatistical modeling. A statistical classifier is built on top of theunderlying verification sources. The classifier combines the multipleverification inputs with associated models of the respective verificationtechnologies underlying the verification inputs. The classifier functions toclassify a claimant as a true speaker or impostor based upon the verificationinputs, the model or models and, potentially, additional information relatedto the context in which the classifier is being used.